Most developmental signaling pathways control gene expression by altering the activity of sequence-specific transcription factors. However, these transcription factors require assistance from chromatin remodeling, histone modification and coactivator or corepressor complexes to regulate transcription. Such complexes have been thought to play a very general, non-specific role in transcription. Surprisingly, mutations in Drosophila genes encoding some subunits of these complexes affect only specific sets of target genes. The aim of this proposal is to understand the basis of this specificity. Osa is an ARID domain protein that is a component of a subset of Brahma chromatin remodeling complexes. Osa negatively regulates genes that normally require Wingless signaling for their expression. One goal of this proposal is to understand how Osa modulates the function of the Brahma complex. Genetic screens will be used to identify dosage-sensitive modifiers of osa overexpression phenotypes and to determine the functions of these genes. BAF180, the subunit found in Brahma complexes that do not contain Osa, will be characterized genetically. Genes regulated by Osa, BAF180 and Brahma will be compared using microarray analysis. Osa will also be targeted to an artificial promoter to study its effects and regulation. skuld and kohtalo encode Drosophila homologues of TRAP240 and TRAP230, the two largest subunits of the mediator complex. Mutations in both genes have identical phenotypes; they do not affect cell growth or survival, but cause loss of expression of a subset of target genes of the Wingless, Notch and Hedgehog pathways. To determine what properties of an enhancer make it dependent on Skuld and Kohtalo, variant enhancers will be analyzed in cell culture and in vivo. A yeast two-hybrid system will be used to look for sequence-specific transcription factors that directly bind to these adaptor subunits. Transcription factors and other proteins involved in Skuld and Kohtalo function will also be identified through a genetic screen. Finally, additional genes encoding likely components of transcriptional regulatory complexes will be analyzed genetically to determine whether they affect the output of specific signaling pathways. This will allow us to understand how each pathway interacts with the general transcriptional machinery.